Excited-state acidity of bifunctional compounds: Part 5 5-(2-hydroxyphenyl)-3-phenyl-1,2,4-oxadiazole and 3-(2-hydroxyphenyl)-5-phenyl-1,2,4-oxadiazole

Article abstract of DOI:10.1039/a702708g

Fluorescence emission and excitation spectra and experimental dipole moments are presented for 5-(2-hydroxyphenyl)-3-phenyl-1,2,4-oxadiazole and 3-(2-hydroxyphenyl)-5-phenyl-1,2,4-oxadiazole. These data indicate that the long-wavelength emission of the fo

Full text of DOI:10.1039/a702708g

View Article Online / Journal Homepage / Table of Contents for this issue
Excited-state acidity of bifunctional compounds
Part 5¤ 5-(2-Hydroxyphenyl)-3-phenyl-1,2,4-oxadiazole and
3-(2-hydroxyphenyl)-5-phenyl-1,2,4-oxadiazole”
Carlos Carvalho,a Ira Brinn,a* Wolfram Baumann,b* Heribert Reisb and Zsolt Nagyb
a L aboratorio de Espectroscopia Resolvida no T empo (L ERT ), Instituto de Qu•mica, Universidade
Federal do Rio de Janeiro, Ilha do Fundao, C. P. 68.563, Rio de Janeiro, R.J. 21.941-Brasil
b Institute of Physical Chemistry, University of Mainz, D-55099 Mainz, Germany
Fluorescence emission and excitation spectra and experimental dipole moments are presented for 5-(2-hydroxyphenyl)-3-phenyl-
1,2,4-oxadiazole and 3-(2-hydroxyphenyl)-5-phenyl-1,2,4-oxadiazole. These data indicate that the long-wavelength emission of the
former, in non-hydrogen bonding solvents, is due to a tautomeric PT structure. The latter exhibits only the “normalÏ band. The
striking di†erence in behavior between these two compounds of very similar structure is explained on the basis of coplanarity
between the aryl ring in position 5 and the heterocycle ring, the aryl ring in the 3 position not being in the same plane.
Excited state intramolecular proton transfer (ESIPT) in
bifunctional compounds dissolved in non-polar solvents has
been a subject of intense interest for the last twenty years, or
so. This interest stems not only from potential applications of
this process to the lasing process,2h4 as a solar energy Ðlter5
and as molecular size6 computer memories, but also from a
more fundamental interest in understanding the e†ect of exci-
tation on electronic redistribution. It is the speed of the
ESIPT process (normally 1 ps or less) which allows it to
monitor this redistribution.
results not only help to explain the ESIPT behavior of these
compounds, but also help to rationalize this observed behav-
ior in terms of molecular conformation.
Methods
3HPPO and 5HPPO were prepared as described27 previously
for the latter. As a basis for comparison, the previously unre-
ported methoxy compound, 5MPPO, toluyl compound,
5HPTO and double phenolyl compound, DHPO were also
prepared. 5MPPO was synthesized by dissolving 20 mg of
5HPPO in 2 ml of diethyl ether and allowing it to react with 6
mg of sodium ethoxide. An excess of methyl iodide was then
added and the solution was stirred for 2 h at room tem-
perature. When the solvent was extracted and the resulting
solid recrystallized from n-hexane, 8 mg (38%) of the product
was obtained; mp \ 122È123 ¡C. 5HPTO was synthesized by
slowly adding a solution of 52 mg of 2-hydroxybenzoyl chlo-
ride in 5 ml of dry pyridine to a previously prepared solution
of 50 mg of 2-methylbenzamidoxime in 10 ml of the same
solvent and reÑuxing for 24 h. The solvent was extracted and
the solid recrystallized from n-hexane, yielding 21 mg (25%)
of solid, in the form of needles; mp \ 108È109 ¡C. DHPO
Compounds containing the heterocyclic ring 1,2,4-oxadia-
zole are important in their own right and three review arti-
cles,7h9 have been devoted to them. Their main applications
arise from their physiologic activities, compounds being
utilized as antitussigens,10,11 coronary vasodilators,12
anticonvulsants,13
hypocholesterics,14
tranquilizers,15
anorexigens,16 anthelmintics,17 antiesquistosoms18 and anti-
microbians.19 However, they also have been used as pesti-
cides,20 herbicides,21 fungicides22 and acaracides23 in
agriculture and antistatic agents,24 thermal stabilizers25 and
blue dyes26 in the polymer industry.
Dual Ñuorescence from 5-(2-hydroxyphenyl)-3-phenyl-1,2,4-
oxadiazole (5HPPO) in non-hydrogen-bonding solvents was
initially explained27 as arising from a “normal, N, conÐgu-
ration (structure 1a in Fig. 1, band at ca. 350 nm) and from a
hydrogen-bonded structure (long wavelength, LW, band at ca.
500 nm.) The experimental evidence cited to argue against the
LW band being due to a proton transfer (PT) structure
(structure 1b in Fig. 1) was the di†erent excitation spectra for
the two bands, indicating that both structures are already
present in the ground state. These same experimental results
were interpreted28 by another research group as being consis-
tent with the LW band being due to a PT structure, the
proton from the hydroxy group being transferred to the
oxygen atom of the heterocycle. This paper reports
four di†erent types of experimental results on 5HPPO,
3-(2-hydroxyphenyl)-5-phenyl-1,2,4-oxadiazole (3HPPO), 5-(2-
methoxyphenyl)-3-phenyl-1,2,4-oxadiazole (5MPPO), 5-(2-
hydroxyphenyl)-3-(2-methylphenyl)-1,2,4-oxadiazole (5HPTO)
and the double phenolyl compound, 3,5-bis(2-hydroxyphenyl)-
1,2,4-oxadiazole (DHPO) which were prepared recently. These
was synthesized by adding
2-hydroxybenzoyl chloride in 5 ml of dry pyridine to a pre-
viously prepared solution of 50 mg of 2-
a solution of 40 mg of
¤ Part 4: ref. 1.
Fig. 1 Possible structures for 5HPPO; 1a normal structure, 1b zwit-
terionic PT structure, 1c ketonic PT structure and 1d PT structure (H
transferred to oxygen)
” Taken, in part, from the MS thesis of C. E. M. Carvelho, Uni-
versidade Federal do Rio de Janeiro, 1996.
J. Chem. Soc., Faraday T rans., 1997, 93(18), 3325È3329
3325

View Article Online
hydroxybenzamidoxime in 10 ml of the same solvent and
reÑuxing for 24 h. The solvent was extracted and the solid
recrystallized from n-hexane, yielding 15 mg (18%) of product;
mp \ 169È171 ¡C. The 1H, 13C NMR, vibrational and mass
spectra were all consistent with the assumed structures.
All of the solvents used for synthesis were reagent grade
(P.A.) from either Merck/Brasil or Grupo Quimica/Brasil. The
solvents used for spectroscopy were acetonitrile, benzene,
chloroform, dichloromethane, n-hexane and N,N-dimethyl-
formamide (Merck-UVASOL) and butan-2-ol, chloroform, n-
hexane and propan-2-ol (Grupo Qu•mica-UV/HPLC). These
were not puriÐed further. The solvents used for the electro-
optical dipole moment determinations were reagent grade
cyclohexane and dioxane. These were puriÐed, as
described29,30 elsewhere and thoroughly dried by reÑuxing in
the presence of a NaÈK liquid alloy.
latter is an nÈp* state while S of 5HPPO and DHPO are
pÈp* states, but, more speciÐcally, that the electronic distribu-
1
tions of the S states are quite distinct in 3HPPO, as com-
1
pared to the other two.
Excitation spectra
For 5HPPO, the three-dimensional excitation-emission
contour diagram shown in Fig. 2 clearly di†erentiates between
the excitation spectrum of the N band and that of the LW
band, in a 1.17 ] 10~4 M n-hexane solution. The longer wave-
length excitation maximum for the N band is located at
300 ^ 2 nm, while a second maximum appears at 270 ^ 2 nm.
The Ðrst excitation maximum for the LW band is located at
323 ^ 2 nm, the second excitation maximum coinciding with
that of the N band. The continuous diagonal lines represent
Raman and elastic scattering, in the upper left-hand corner,
and second-order scattering in the lower right-hand corner of
the contour diagram. This suggests that the ground-state
Fluorescence spectra were taken using an Hitachi F-4500
Ñuorescence spectrophotometer. The electro-optical absorp-
tion measurements were made on an instrument
described30,31 previously.
Results and Discussion
Fluorescence spectra
The Ñuorescence maxima of 3HPPO, 5HPPO, DHPO,
5HPTO and 5MPPO in solvents of di†erent polarity are
listed in Table 1. For 3HPPO, only the N band is observed in
both polar and non-polar solvents. 5HPPO is quite di†erent,
showing an N band in both polar and non-polar solvents as
well as a broad LW band in the same non-polar solvents.
5MPPO, which cannot undergo proton transfer, is similar to
5HPPO, only without the LW band. DHPO and 5HPTO are
also quite similar to 5HPPO, showing two Ñuorescence bands
in a non-polar solvent, but only one band in a polar solvent,
whose maxima appear at approximately the same wavelength
as 5HPPO. For both DHPO and 5HPTO the interpretation
that can be given is that neither the addition of a methyl
group nor a hydroxy group in an ortho position on the ring in
the 3-position has an appreciable e†ect on the emission spec-
trum of 5HPPO. In addition, for 5HPPO in non-polar sol-
vents, the ratio of the intensities of the two emission peaks
depends on excitation wavelength (in a non-linear fashion)
and the extent to which the solvent has been dried by reÑux-
ing over Na (the drier the solvent the greater the contribution
of the LW band).
The bathochromic shifts of 15 nm for the N band of
5HPPO and 6 nm for the same band of DHPO as the solvent
polarity increases contrasts with the corresponding hypso-
chromic shift of 16 nm for the N band of 3HPPO. This not
only suggests that the Ðrst excited singlet state (S ) of the
Fig. 2 Contour diagram of 5HPPO. j vs. j
ex
em
1
Table 1 Fluorescence emission maximaa (nm)
solvent
n-hexane
benzene
*f
3HPPO
5HPPO
DHPO
5MPPO
345
5HPTO
0.001
0.0039
0.1408
345
498
346
498
350
498
354
500
348
500
chloroform
350
348
dichloromethane
dichloroethane
0.2183
0.223
356
498
359
butan-2-ol
propan-2-ol
dimethylformamide
acetonitrile
0.2606
0.2744
0.2755
0.3062
333
360
348
355
366
377
498
a All values ^ 0.5 nm.
3326
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
structures responsible for the two bands are appreciably dif-
ferent, that our previous interpretation27 is incorrect and that
these experiments can be interpreted as conÐrming the assign-
ment of the N band, favored by the presence of small quan-
tities of water in non-polar solution and indicating that the
LW band is due to a PT structure whose origin is a pre-
where
s(/) 4 3 cos2 / [ 1
(3)
(4)
(5)
(6)
r(/) 4 2 [ cos2 /
t(j) 4 ([j/5hci)[d(ij)/dj]
u(j) 4 (j2/10h2c2i)[d2(ij)/dj2]
existent hydrogen bond in the ground state (S ).
0
This interpretation leads to the conclusion that 3HPPO
/ is the angle between the external Ðeld E and the electric Ðeld
vector of the linearly polarized light used in the experiment
(usually the values of / \ 0 and 90 are used) i(j, /) is the
absorption coefficient, and t(j) and u(j) are related to the Ðrst
and second derivatives of the absorption band with respect to
wavelength, respectively, and can be determined30,31,34 inde-
pendently from the absorption spectrum.
does not exhibit an LW band, in part, because it does not
establish a viable hydrogen bond in S . This conclusion is
0
supported by the 1H NMR spectra measured in CDCl . The
3
hydroxy proton in 3HPPO exhibits a chemical shift of only d
9.62, whereas the corresponding value for 5HPPO is d 10.5,
indicating the latter to be more acidic. The double phenolyl
compound, DHPO, shows values of d 8.65 and 10.3, consis-
tent with its having one proton of each type. In addition, it is
worthwhile to note that all of the 1H NMR hydroxy peaks are
extremely sharp, indicating the presence of only one con-
The coefficients E through I in eqn. (2) are deÐned, in non-
polar solvents, as follows
E 4 b2f 2[3(ml )2 [ l
–
2]
(7)
(8)
e
g
g
former. This interpretation is also consistent with the pK
a
values in aqueous solution, 9.03 ^ 0.05 for 5HPPO compared
F 4 b2f 2 l *l
e
g
to 9.8 ^ 0.1 for 3HPPO. These values not only show the
hydroxy group on the ring in the 5-position to be consider-
ably more acidic than the hydroxy group on the ring in the
3-position, but also indicate that the heterocyclic ring has
G 4 b2f 2(ml )(m *l)
(9)
e
g
H \ f 2(*l
–
)2
(10)
(11)
e
I \ f 2(m *l)2
–
little e†ect on acid dissociation in 3HPPO, since the pK of
e
a
3HPPO is quite close to the corresponding value32 for
b 4 (1/kT ), m is a unit vector in the direction of the moment
phenol, i.e., 10.0 Combining the evidence of the emission
spectra with that of the excitation spectra leads one to suggest
that the aryl ring in the 5-position is coplanar with the hetero-
cylic ring whereas the aryl ring in the 3-position is not. Thus,
for 3HPPO, not only is there no pre-established hydrogen
bond in the ground state (lower d value) but also no ESIPT is
observed because the hydroxyphenyl group in the 3-position
is not involved in the excitation, being considerably out of the
plane of the other two rings. It is worthwhile emphasizing that
it is the presence of the aryl ring in the 5-position in 3HPPO
that forces the hydroxyphenyl ring in the 3-position out of
coplanarity. This is illustrated by the fact that 3-(2@-hydroxy-
phenyl)-5-methyl-1,2,4-oxadiazole, which can be considered to
be 3HPPO with a methyl group substituted in the 5-position
instead of a phenyl group, exhibits33 dual Ñuorescence in non-
polar solvents and therefore has coplanar aryl and hetero-
cyclic rings.
of the transition being considered, l is the total dipole
g
moment of the ground state, l the total dipole moment of the
e
FranckÈCondon excited state and *l 4 (l [ l ) in the case
e
g
of non-polar solvents. Total dipole moment here means that
some amount of solvent induced moment (usually, around
10%) is included. The separation of permanent and induced
moments can be done on the basis of the Onsager model
(these details will not be discussed here), f is a Ðeld correction
e
factor29h31,34,35 whose value is 1.2 for non-polar solvents with
relative permittivity ca. 2.
Fig. 3 shows a plot of experimental L values vs. j, between
304 and 340 nm for 5HPPO. A multilinear regression analysis
according to eqn. (2) (using / \ 0 and 90¡ and j at 4 nm
intervals) yields the coefficients EÈI. These results are shown
in Table 2 for both 5HPPO and 3HPPO; for the former the
regression coefficient r was 0.991 whereas for 3HPPO the cor-
responding value was 0.992, i.e. both are considered to be
Electro-optical dipole moment determination
From the e†ect of an external electric Ðeld (E) on the absorp-
tion and Ñuorescence spectra of compounds in solution the
dipole moments of the respective ground and excited states
can be determined.31 In addition, this method generates infor-
mation about the homogeneity34 of the bands observed.
Although the Ñuorescence bands of 3HPPO and 5HPPO were
too weak to apply this method, an analysis of the Ðrst absorp-
tion band was performed. This analysis is capable of deter-
mining the ground state and FranckÈCondon excited singlet
state dipole moments. In the case of DHPO it was not pos-
sible to perform the same analysis, apparently because the Ðrst
band is not a pure transition.
Fig. 3 L vs. j for 5HPPO
The method generates a macroscopic quantity (L ), Ðrst
deÐned35 by Liptay as
Table 2 Regression coefficients and electro-optical dipole moment
determination
L (j, /) 4 s(j, /)/E2
(1)
coefÐcient [eqn. (6)]
5HPPO
3HPPO
where s is the relative Ðeld induced change of the absorption
coefficient and j is wavelength. L can be related30,31,34 to
microscopic quantities by
E/10~20 m2 V~2
F/10~40 C V~1 m2
G/10~40 C V~1 m2
H/10~30 C2 m2
56 ^ 8
38 ^ 3
1630 ^ 70
[500 ^ 30
[590 ^ 30
310 ^ 70
(310)
44 ^ 3
L (j, /) 4 (E/30)s(/) ] [Fr(/) ] Gs(/)]t(j)
145 ^ 9
144 ^ 9
I/10~30 C2 m2
] [Hr(/) ] Is(/)]u(j)
(2)
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
3327

View Article Online
excellent Ðts. This conÐrms the hypothesis that only one con-
formational species is absorbing over the entire wavelength
range studied in both compounds.
For both 3HPPO and 5HPPO, the results show that F B G
and H B I, to a good approximation. This means that m is
parallel to l and to *l, within experimental error. Starting
g
from this observation, eqn. (7)È(11) can be used to derive
values for l , l *l and l and results are shown in Table 3.
g
g
e
This allows the calculation of l , which turns out to be paral-
e
lel to l for 5HPPO but antiparallel for 3HPPO, as shown in
g
Table 3. It is worth noting that o k o [ o k o for 5HPPO,
e
g
whereas o k o [ o k o for 3HPPO, consistent with the spectro-
g
e
scopic results and most likely indicating that S for 5HPPO is
1
a pÈp* state whereas it is an nÈp* state for 3HPPO. It is also
worthwhile to point out the good agreement of k , indepen-
g
Fig. 4 Solvatochromic plot for 5HPPO
dent of the variables used to obtain results for both com-
pounds.
approximation for this compound.) This small di†erence
between the dipole moment values for the Ñuorescent state
and the ground state of the tautomer suggest that the PT
structure is not very polar and point to a structure which is
considerably closer to the ketonic structure 1c than to the
zwitterionic structure 1b (Fig. 1), as being responsible for the
LW band. These values also weigh strongly against structure
1d, which was previously suggested,28 because no reasonable,
strictly covalent, structure can be drawn for it.
Solvatochromic dipole moment determination
Utilizing the maxima of the absorption and emission bands as
a function of solvent polarity, Lippert derived36,37
(k [ k )2 \ *k2 \ hca3*l/2 *f
(12)
e
g
where
*f 4 (D [ 1)/(2D ] 1) [ (n2 [ 1)/(2n2 ] 1)
(13)
*f is a measure of solvent polarity, D is the relative permit-
tivity and n is the refractive index of the solvent, *l is the
average of the maxima of the absorption and emission bands
(in cm~1), and a is the “e†ectiveÏ spherical radius of one solute
molecule, which is taken as
Conclusions
It is suggested that the LW emission observed for 5HPPO is
from the covalent, ketonic, PT structure (1c in Fig. 1). It is
further suggested that the di†erence in excitation spectra of
the two structures in 5HPPO is due to the presence of mois-
ture in the non-polar solvent, which favors the N emission.
The LW band is not observed for 3HPPO, which is attributed
to the aryl ring in the 3-position being considerably out of the
plane shared by the other two rings.
a + M3M/4noNN1@3
(14)
where M is the molecular weight and o is the density of the
solute, N is AvogadroÏs number, h is PlanckÏs constant and c,
the velocity of light. A plot of *l vs. *f should yield a straight
line of slope \ 2 *k2/hca3, which yields k , if k is known.
The authors gladly acknowledge the World Bank and the
Funding Agency for Studies and Projects (FINEP) for an
equipment grant, the Jose Bonifacio University Foundation
(FUJB) for a maintenance grant, the Fonds der Chemischen
Industrie, Germany (to W.B.) and the Brazilian National
Research Council (CNPq) for partial Ðnancial support (to
C.E.M.C. and I.M.B.).
e
g
Using the average of the maxima of the absorption and
emission (taken from Table 1) of the N band of 5HPPO as a
function of solvent polarity, in n-hexane, benzene, chloroform,
dichloroethane, butan-2-ol, N,N-dimethylformamide and ace-
tonitrile yielded k \ (19 ^ 3) ] 10~30 C m (see Fig. 4, multi-
e
ple r \ 0.909) for the N state, in fair agreement with the k
e
value of 11.8 ] 10~30 C m for the FranckÈCondon state
determined by the electro-optical absorption method. This
suggests that there is no large change in structure, nor elec-
tronic distribution, upon relaxation of the structure
responsible for the N emission, which is not surprising for
what is expected to be a fairly rigid, monomeric, molecular
structure. When applied to the LW emission of 5HPPO, the
solvatochromic results obtained with the same solvents, with
the exception of butan-2-ol and N,N-dimethylformamide,
yielded *k B 0, as can be seen from inspection of Table 1.
Although the value for *k in the case of the N band is sensi-
tive to the e†ective spherical Onsager radius chosen, the latter
value, of course, is not. (A value of 3.85 Ó was used for this
radius, based on the assumptions of a density of 1.0 g cm~3
and efficient packing. However, 5HPPO is not expected to be
spherical, therefore the Onsager treatment is only a rough
References
1
2
3
4
T. D. S. Tseluiko and I. M. Brinn, J. Photochem. Photobiol A,
1994, 84, 139.
A. U. Khan and M. Kasha, Proc. Natl Acad. Sci., USA, 1983, 80,
1767.
P.-T. Chou, D. McMorrow, T. J. Aartsma and M. Kasha, J.
Phys. Chem., 1984, 88, 4596.
A. Costela, J. M. Munoz, A. Douhal, J. M. Figuera and A. U.
Acuna, Appl. Phys. B, 1989, 49, 545.
N. A. Shaath, Cosmetics T oiletries, 1987, 21, 102.
S. Esener and P. M. Rentzepis, Proc. SPIEÈInt. Soc. Opt. Eng.,
1991, 1499, 144. (Chem. Abstr., 1991, 115, 194057k).
L. C. Behr, T he Chemistry of Heterocyclic Compounds ed. A.
Weissberger, Wiley-Interscience, New York, 1962, p. 245.
F. Eloy, Fortschr. Chem. Forsch., 1965, 4, 807.
L. B. Clapp, Adv. Heterocycl. Chem., 1976, 20, 65.
5
6
7
8
9
10 M. De Gregorio, Panminerva Med., 1962, 4, 90.
11 K. Harsanyi, P. Kiss, D. Korbonits and I. R. Malyata, Arzneim-
Forsch., 1966, 16, 615.
Table 3 Electro-optical dipole moments for the N structure
dipole moment
5HPPO
3HPPO
variables used
12 F. Eloy and R. Lenaers, Bull. Chem. T herap., 1966, 347.
13 M. Nakanishi, K. Araki, T. Tahara and M. Shiroki, Ger. Pat.,
2 331 540, 1974 (Chem. Abstr., 1974, 80, P96036s).
14 Y. Imai and K. Shimamoto, Atherosclerososis, 1973, 17, 131.
15 B. Cavalleri and G. Lancini, Ger. Pat., 2 331 058, 1974 (Chem.
Abstr., 1974, 80, P95961c).
o *k o/10~30 C m
10.0 ^ 0.4
11.8 ^ 1.6
1.8 ^ 0.2
1.2 ^ 0.2
11.8 ^ 0.5
k CCk
14.7 ^ 1.5
[156 ^ 20
9.8 ^ 0.2
10.6 ^ 1.7
4.9 ^ 1.5
k CBk
H, I
F, G
k *k/10~60 C2 m2
g
o k o/10~30 C m
E
g
o k o/1030 C m
F, G, H, I
E, H, I
g
o k o/10~30 C m
e
16 R. D. Vedova and G. DÏAlo, Bull. Chim. Farm., 1973, 112, 273
(Chem. Abstr., 1974, 80, 10431g).
e
g
e
g
3328
J. Chem. Soc., Faraday T rans., 1997, V ol. 93

View Article Online
17 C. Ainsworth, W. E. Buting, J. Davenort, E. M. Callender and M.
C. McCowen, J. Med. Chem., 1967, 10, 208.
27 I. M. Brinn, C. E. M. Carvalho, F. Heisel and J. A. Miehe, J.
Phys. Chem., 1991, 95, 6540.
18 C. H. Robinson, E. Duelding and J. Fisher, Mol. Pharmacol.,
1970, 6, 604.
28 P.-T. Chou and M. L. Martinez, J. Phys. Chem., 1992, 96, 2018.
29 T echniques of Organic Chemistry, Organic Solvents, ed. A. Weiss-
berger, Interscience New York, 2nd edn, 1955, vol. 7.
30 W. Baumann, Physical Methods of Chemistry, 2nd edn., vol. III;
Determination of Chemical Composition and Molecular Structure,
Part B, ed. B. W. Rossiter and J. F. Hamilton, J. Wiley and Sons,
New York, 1989.
19 A. J. Crovetti, A. M. Von Esch and R. J. Thill, J. Heterocycl.
Chem., 1972, 9, 435.
20 P. E. Keeley, C. H. Carter and J. H. Miller, W eed Sci., 1973, 21
327 (Chem. Abstr., 1973, 79, 112306s).
21 H. Moser and J. Rumpf, S. Afr. Pat. 6706 664, 1968 (Chem. Abstr.,
1969, 70, P77976a.)
31 W. Rettig and W. Baumann, Prog. Photochem. Photophys, ed. J.
F. Rabek, CRC Press, Boca Raton, FL, vol. 6, 1990.
32 A. I. Biggs and R. A. Robinson, J. Chem. Soc., 1961, 388.
33 A. M. G. Pereira, I. M. Brinn, R. M. Srivastava and L. Martin, J.
Chem. Soc., Faraday T rans. 2, 1984, 80, 763.
34 W. Liptay, Excited States, ed. E. C. Lim, Academic Press, New
York, 1974. vol. 1.
35 W. Liptay, Z. Naturforsch., T eil A, 1965, 20, 272.
36 E. Lippert, Z. Naturforsch, T eil A, 1955, 10, 541.
37 E. Lippert, Z. Elektrochem., 1957, 61 962.
22 H. Hamano, K. Shimada and S. Kuriyama, Jpn. Pat., 7 203 832,
1972 (Chem. Abstr., 1972, 76, 153752v).
23 J. Wood and M. J. Bull, Ger. Pat., 2 312 500, 1973 (Chem. Abstr.,
1973, 79, P146504v).
24 H. Hagen, F. Becke and J. Niemeyer, (a) Ger. Pat., 2 016 696,
1971 (Chem. Abstr., 1972, 76, P25299m); US Pat. 3 776 910, 1973
(Chem. Abstr., 1974, 80, P82995s).
25 D. A. Bochvar, I. V. Stankevich, E. S. Kronganz and V. V.
Dorshak, Dokl. Akad. Nauk., SSSR, 1971, 200, 364 (Chem. Abstr.,
1972, 76, 46591y).
26 M. Patsch and H. Eilingsfeld, Ger. Pat., 1 949 295, 1971 (Chem.
Abstr., 1971, 75, P7417y).
Paper 7/02708G; Received 21st April, 1997
J. Chem. Soc., Faraday T rans., 1997, V ol. 93
3329